Skip to main content

Table 1 Misclassification rate - case 1

From: Comparison of classification methods that combine clinical data and high-dimensional mass spectrometry data

 

μ =0 . 5

 

n=100

n=200

n=500

1) Clin

28 (11)

29 (6)

27 (4)

2) Boost

46 (11)

45 (7)

37 (5)

3) BoostOffClin

30 (11)

27 (7)

18 (3)

4) ClinOffBoost

32 (11)

27 (5)

22 (4)

5) SGPLS

49 (11)

45 (8)

38 (5)

6) SGPLSOffClin

44 (13)

35 (12)

24 (5)

7) ClinOffSGPLS

32 (11)

28 (6)

23 (4)

8) RPLS

47 (12)

43 (7)

40 (4)

9) RPLSOffClin

27 (9)

27 (6)

24 (5)

10) ClinOffRPLS

39 (11)

37 (8)

34 (5)

Oracle

18 (8)

16 (5)

13 (3)

  1. This table contains the mean and the standard deviation of the MCR over 50 simulated datasets for each compared method and for μ=0.5 (i.e. lower predictive importance of MS data than clinical data).
Back to article page

BMC Bioinformatics

ISSN: 1471-2105

Contact us